US 12521400

Summary — U.S. Patent No. 12,521,400 (US12521400B2)

Bibliographic data (confirmed by Google Patents record and third-party USPTO-derived sources)

Abstract (verbatim from the record): "The invention relates to pharmaceutical compositions comprising one or more benzodiazepine drugs for nasal administration, methods for producing and for using such compositions."

Prosecution / assignment history: The application was assigned by the inventors (and others including David F. Hale and Edward T. Maggio) to Hale BioPharma Ventures, LLC on July 24, 2024, and then to Neurelis, Inc. on the same date. The granted patent's inventor list per Google Patents is the five names above.

Plain-language overview of the invention

The patent is one of Neurelis's family of patents covering VALTOCO®, a rescue treatment for seizure clusters. The specification explains that oral benzodiazepines act too slowly and suffer first-pass liver metabolism; IV administration requires medical personnel; and rectal suppositories are impractical for caregivers. The invention solves this by formulating a benzodiazepine (chiefly diazepam) as a stable, non-aqueous nasal spray solution in which the drug is dissolved in a carrier made of vitamin E (tocopherols/tocotrienols) plus a lower alcohol or glycol (e.g., ethanol, benzyl alcohol, propylene glycol), optionally with an alkyl-glycoside penetration enhancer (e.g., dodecyl maltoside). The nasal route gives rapid absorption and fast onset (minutes) so a caregiver can stop a seizure without needles or suppositories.

Two broad embodiments are described: (1) solution formulations where the drug is fully dissolved and substantially free of water and of drug microparticles/nanoparticles, and (2) particulate (micro-/nanoparticle suspension) formulations (the specification includes "consisting of" embodiments built around Vitamin E TPGS, parabens, propylene glycol, povidone, optionally alkyl glycoside, and water). Note that several passages expressly disclaim Vitamin E TPGS and polymeric glycols in the solution embodiments.

Independent claims — overview with stated uncertainty

Important caveat: The full claims text was not included in the copy of the patent supplied to me, and I have not verified the complete claim set directly on the USPTO Patent Center. The following is based on the specification plus the description of the asserted claims in the Padagis ANDA complaint (D. Del. 1:26-cv-00401), which states Neurelis asserts claims 1–21 and focuses on independent claim 1. I cannot confirm with high confidence how many independent claims exist or whether any are method-of-treatment claims, so treat the claim details below as preliminary.

Independent claim 1 (composition claim) — as characterized in the complaint:
A pharmaceutical composition for intranasal administration containing:

• about 1 to about 20 mg of diazepam (or a pharmaceutically acceptable salt) fully dissolved in a carrier;
• about 45% to about 85% (w/w) of one or more natural or synthetic tocopherols or tocotrienols selected from a defined group (α/β/γ/δ-tocopherol, α/β/γ/δ-tocotrienol, tocophersolan, isomers, esters, combinations);
• about 25% to about 40% (w/w) of one or more alcohols selected from a defined group (e.g., ethanol, propyl/butyl/pentyl/benzyl alcohol);
• wherein the composition is a solution of 50 µL to 150 µL presented in a pharmaceutically acceptable nasal spray formulation for intranasal administration.

In plain language: A single-dose (50–150 µL) nasal-spray solution containing a 1–20 mg therapeutic dose of diazepam dissolved in a vitamin-E–type carrier (45–85% by weight) mixed with an alcohol solvent (25–40% by weight), formulated for spraying into the nose to rapidly treat seizures.

Likely additional independent claims (inferred, not verified): Given the specification's repeated "the invention also discloses a method of treating a patient… comprising administering to one or more nasal mucosal membranes" language and the "consisting of"/"consists essentially of" solution embodiments, the claim set very likely also includes (a) one or more method-of-treatment claims reciting intranasal administration of the diazepam solution (with dosing/spraying steps such as spraying into one or both nostrils, optionally with a pre-selected delay between doses) and possibly (b) narrower "consisting of" composition claims. I cannot verify the exact wording or numbering without the issued claims.

Litigation / CAFC docket status

CAFC (Federal Circuit) 2026 dockets: My searches of CAFC dockets for "12521400" / "12,521,400" returned no results. That is consistent with timing: the patent only issued January 13, 2026, and the pending cases are at the district-court/PTAB stage, so no appeal to the Federal Circuit involving this specific patent number would be expected yet as of the search date.

District court (D. Del.) ANDA litigation naming the '400 patent (per CourtListener / Justia / Unified Patents data):

• Neurelis, Inc. v. Padagis LLC et al., 1:26-cv-00401 (D. Del.) — filed April 8, 2026, after Padagis's March 12, 2026 Paragraph IV notice letter; asserts claims 1–21 of the '400 patent against a generic VALTOCO (diazepam nasal spray 10 mg) ANDA product. Consolidated with lead case 24-562 (MN) before Judge Noreika; Markman hearing set for August 19, 2026.
• Neurelis, Inc. v. Lupin Inc. et al., 1:26-cv-00258 (D. Del.) — filed March 11, 2026; report to the Commissioner lists Patent No. 12,521,400; 30-month stay deadline noted as Nov. 27, 2027.
• Neurelis, Inc. v. Amneal, 1:26-cv-00261 (D. Del.) — also listed in Google Patents' litigation links for this patent.
• Neurelis, Inc. v. Strides Pharma Global PTE Ltd., 1:26-cv-00947 (D. Del.) — filed July 31, 2026 (after my nominal search date of April 26, 2026, so this is from later-indexed sources; it also lists the '400 patent among asserted patents).
• The Google Patents record flags this patent's family as having "First worldwide family litigation" and lists the Delaware cases above.

PTAB context (not against the '400 patent itself): Padagis has filed IPRs (e.g., IPR2025-00464, -00465, -00466) against related family patents (8,895,546; 11,241,414; 11,793,786); institution was granted with trials scheduled. No IPR specifically against the '400 patent was identified in my searches.

Confidence notes / uncertainties

• High confidence: Title, assignee, inventors, filing/grant dates, abstract, Orange Book association with VALTOCO, and the existence of the Delaware ANDA actions listed above.
• Moderate confidence: The precise wording of independent claim 1 (taken from the complaint's characterization, which cites col. 37 of the patent).
• Low / unverified: The complete claim set — total number of claims, whether additional independent method claims exist, and their exact text. The claims portion was not in the text provided and I could not access the USPTO Patent Center claim viewer within my search step limit. If you need claim-by-claim detail, the authoritative source is the USPTO Patent Center record for the '400 patent or the complaint exhibits in 1:26-cv-00401.
• No CAFC docket entries for this patent number were found; if you are tracking appeals, re-check the Federal Circuit docket after the district-court cases reach judgment.

Based on the search results available (which include the Google Patents litigation flags for this patent family and PACER/docket-derived records), here is what is known about litigation involving US Patent No. 12,521,400 ("Administration of benzodiazepine compositions" — the VALTOCO® diazepam nasal spray patent, issued January 13, 2026, assigned to Neurelis, Inc.). Note: I could not complete additional live verification of every docket (search step limit reached), so statuses below reflect the sources cited and may not be fully current.

Cases specifically asserting US 12,521,400

1. Neurelis, Inc. v. Lupin Inc. et al. — 1:26-cv-00258 (D. Del.)

• Plaintiff: Neurelis, Inc.
• Defendants: Lupin Inc., Lupin Limited, Lupin Pharmaceuticals, Inc.
• Jurisdiction: U.S. District Court for the District of Delaware (Judge Maryellen Noreika)
• Case number: 1:26-cv-00258 (also styled 1:2026cv00258)
• Filing date: March 11, 2026
• Type: Hatch-Waxman ANDA patent case (Nature of Suit 835); 35 U.S.C. § 271
• Key details: The docket's Patent/Trademark Report identifies patent number 12,521,400; ANDA notice received October 17, 2025; patent expiration date March 27, 2029; thirty-month stay deadline November 27, 2027. Lupin entities filed an answer with affirmative defenses and a counterclaim against Neurelis.
• Status: Pending (per docket entries through the case-activity record; no final judgment found). Case-related filings reference consolidation into lead case C.A. No. 24-562 (MN).
• Sources: pacermonitor.com/public/case/63524746; dockets.justia.com/docket/delaware/dedce/1:2026cv00258/92354

2. Neurelis, Inc. v. [Amneal Pharmaceuticals LLC](/litigations/by-plaintiff/Amneal%20Pharmaceuticals%20LLC) et al. — 1:26-cv-00261 (D. Del.)

• Plaintiff: Neurelis, Inc.
• Defendants: Amneal Pharmaceuticals LLC and Amneal Pharmaceuticals, Inc.
• Jurisdiction: U.S. District Court for the District of Delaware (Judge Maryellen Noreika)
• Case number: 1:26-cv-00261 (also styled 1:2026cv00261)
• Filing date: March 11, 2026
• Type: Hatch-Waxman ANDA patent case (Nature of Suit 835); 35 U.S.C. § 271
• Key details: Complaint asserts claims 1–21 of the '400 patent (along with other Neurelis patents) against Amneal's ANDA No. 219109 (generic VALTOCO®, 7.5 mg and 10 mg diazepam nasal sprays). Amneal answered. A Second Consolidation Order and Modified Case Scheduling Order (dated April 23, 2026) consolidated the matter into lead case C.A. No. 24-562 (MN), with a Markman hearing set for August 19, 2026 and a bench trial set for the week of January 10, 2028.
• Status: Pending, consolidated into lead case C.A. No. 24-562 (MN).
• Sources: pacermonitor.com case 63525453; ai-lab.exparte.com/case/dct/ded/1:26-cv-00261; open-public-records.com

3. Neurelis, Inc. v. Padagis LLC et al. — 1:26-cv-00401 (D. Del.)

• Plaintiff: Neurelis, Inc.
• Defendants: Padagis LLC, Padagis US LLC, and Padagis Israel Pharmaceuticals Ltd.
• Jurisdiction: U.S. District Court for the District of Delaware (Judge Maryellen Noreika)
• Case number: 1:26-cv-00401 (styled 1:26-cv-00401-MN)
• Filing date: April 8, 2026
• Type: Hatch-Waxman ANDA patent case (Nature of Suit 835); 35 U.S.C. § 271
• Key details: This is the "Padagis IV" action arising from Padagis's fourth notice letter (March 12, 2026) specifically regarding U.S. Patent No. 12,521,400 (which issued January 13, 2026). Padagis ANDA No. 219320 is a generic version of VALTOCO®. Claims 1–21 of the '400 patent are asserted. The complaint describes it as related to the Padagis I, II, III, Lupin I and II, and Amneal I actions. ANDA notice received March 12, 2026; expiration March 27, 2029. Consolidated by Second Consolidation Order (April 23, 2026) into lead case C.A. No. 24-562 (MN).
• Status: Pending, consolidated into lead case C.A. No. 24-562 (MN); answer due April 30, 2026.
• Sources: courtlistener.com/docket/73167119; ai-lab.exparte.com/case/dct/ded/1:26-cv-00401; drugpatentwatch.com

4. Neurelis, Inc. v. Strides Pharma Global PTE Limited — 1:26-cv-00947 (D. Del.)

• Plaintiff: Neurelis, Inc.
• Defendant: Strides Pharma Global PTE Limited
• Jurisdiction: U.S. District Court for the District of Delaware (Judge Maryellen Noreika)
• Case number: 1:26-cv-00947
• Filing date: July 31, 2026
• Type: 35 U.S.C. § 271 patent infringement (drug patent litigation)
• Key details: DrugPatentWatch lists U.S. Patent No. 12,521,400 among the seven patents-in-suit (with 11,241,414; 11,793,786; 12,268,664; 12,324,852; 12,337,061; and 12,599,611). No termination date listed as of the record's last update.
• Status: Pending (no disposition found).
• Source: drugpatentwatch.com litigation record for Neurelis, Inc. v. Strides Pharma Global PTE Limited

Notes and caveats

• Not all "related" Neurelis cases assert the '400 patent. The complaint in the Padagis case references earlier related actions ("Padagis I" (filed ~May 8, 2024), "Padagis II" (July 2, 2025), "Padagis III" (October 3, 2025), and "Lupin I" (November 26, 2025)). Because the '400 patent did not issue until January 13, 2026, those earlier-filed cases would not have asserted it; only the post-issuance actions (Lupin II = 1:26-cv-00258, Amneal I = 1:26-cv-00261, Padagis IV = 1:26-cv-00401, and the Strides case = 1:26-cv-00947) name the '400 patent.
• Consolidation: Several of these cases are consolidated for pretrial purposes into lead case C.A. No. 24-562 (MN) (D. Del.), before Judge Maryellen Noreika.
• Google Patents' own litigation flags for this patent point to the Delaware cases 1:26-cv-00258, 1:26-cv-00261, and 1:26-cv-00401, consistent with the above.
• I found no PTAB/IPR proceeding specifically against the '400 patent in the available search results, and I could not confirm any final judgments or settlements for these cases. For fully current docket status (answers, motions, consolidation orders, trial dates), PACER (D. Del.) and the Unified Patents litigation portal should be consulted directly.

I'll verify the structured "no PTAB activity" signal against live sources before concluding.

The structured block says no PTAB activity. Let me run targeted confirmatory searches before I commit to that conclusion.

No post-grant proceeding against the '400 patent surfaced. One critical statutory point needs verification — the § 311(c) IPR start date — plus the family IPR outcomes for context.

PTAB / AIA Trial Report — U.S. Patent No. 12,521,400

Verification basis: canonical structured block ("Patent Trial and Appeal Board proceedings on file") + four targeted web searches (DrugPatentWatch patent-level PTAB page, CourtListener dockets, PTAB Litigation Blog / Director decision documents, Unified Patents Q1 2026 dispute report). I did not have live API access to PTAB E2E or the USPTO ODP in this session, so the negative result rests on the structured block plus published aggregators. Flagged accordingly below.

Proceedings overview

Total AIA trial proceedings naming US 12,521,400: zero (0). Breakdown: 0 active · 0 claims invalidated · 0 claims sustained · 0 settled · 0 institution denied — and, critically, this is a structural zero, not merely an evidentiary one: the IPR window on this patent had not yet opened as of 2026-09-10.

That is the single most important fact for a defendant. Under 35 U.S.C. § 311(c), an IPR petition "may not be filed until after the later of (1) the date that is 9 months after the grant of the patent; or (2) if a post-grant review is instituted, the date of the termination of such post-grant review." The '400 patent granted 2026-01-13, so the earliest date an IPR could be filed is 2026-10-13 (assuming no PGR is instituted). (This is a statutory-text point from my own legal knowledge, not a search result — verify against § 311(c) before relying on it, but it is not close to the line.) No IPR against this patent can exist yet, so the absence of proceedings tells you nothing whatever about whether the claims are vulnerable or whether competitors intend to attack them. It tells you the clock hasn't started.

Bottom-line defensive posture: the patent is not "hardened" and it is not "dead" — it is untested. Claim 1 and all 21 asserted claims are 100% live and 100% un-adjudicated at the PTAB. Any demand letter or complaint invoking claims 1–21 can still be met with a full menu of grounds, and — unusually — there is a narrow remaining window for § 112 grounds via PGR (see below) that IPR can never provide. The defensive posture is "uncontested but vulnerable on a known family-wide issue," and the family-wide issue is not speculative: a sister patent in this same family was invalidated at the PTAB and affirmed on appeal.

For each proceeding

None exist for U.S. 12,521,400. Per the task constraint (don't fabricate proceeding numbers), I am not going to populate the template with invented IPR2026-xxxxx numbers. The two obvious candidate number ranges — IPR2026-003xx and IPR2026-004xx — appear in unrelated matters (e.g., IPR2026-00298, Microsoft v. Qomplx, on U.S. 12,143,424; IPR2026-00204 series, Tesla v. Bulletproof), and nothing in the record ties any of them to this patent.

Adjacent proceedings that DO exist — and precisely why they are not PTAB history of this patent

These matter strategically, but do not cite them as a PTAB record on the '400 patent. If you tell a judge or the Board that "the '400 patent has been through IPR," you are wrong.

IPR2019-00451 — Aquestive Therapeutics v. Neurelis, Inc.

• Type: Inter Partes Review
• Patent challenged: U.S. 9,763,876 (the '876 patent — sibling, not the '400)
• Status: Final Written Decision — claims held unpatentable; affirmed by the Federal Circuit (per the Director's decision text; I did not verify the CAFC docket number and will not guess it)
• Panel/grounds: Grounds were § 103 obviousness over Gwozdz (WO 2009/120933) in view of Meezan (US 2006/0046962); the panel additionally held the challenged claims not entitled to the March 28, 2008 priority date of provisional 61/040,558 (the "'558 provisional")
• Why it matters here: this is the origin of the "material error by the Office" theory that unlocked the current family IPRs, and it is the template for any attack on the '400. Source: https://www.ptablitigationblog.com/informative-no-discretional-denial-where-prior-ipr-suggests-material-error-by-office/ and the PTAB record excerpt at https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1557269](/patent/1557269)

IPR2025-00464 / IPR2025-00465 / IPR2025-00466 — Padagis US LLC v. Neurelis, Inc.

• Type: Inter Partes Review (three parallel petitions)
• Patents challenged: U.S. 8,895,546 (the '546; IPR2025-00464), 11,241,414 (the '414; IPR2025-00465), 11,793,786 (the '786; IPR2025-00466) — all siblings; none is the '400
• Filed: 2025 (petition numbers are IPR2025-series)
• Status: Instituted and in trial. Neurelis's request for discretionary denial was denied by the Acting Director (decision designated informative on 2026-01-09); the Director referred the petitions to the Board, which instituted. Projected Final Written Decision due date: 2026-09-17 (per the Director's decision). A bench trial had been set for 2026-02-09.
• Petition grounds: § 103 obviousness over Gwozdz PCT + Meezan, targeting the same core issues: (a) entitlement to the '558 provisional priority date, and (b) motivation to combine, with Padagis asserting collateral estoppel from IPR2019-00451.
• Panel: the institution decision was directed by Acting Director Coke Morgan Stewart; APJ names for the merits panel were not captured in my sources.
• Settlement / appeal: none identified. No CAFC appeal can exist yet — no FWD had issued as of 2026-09-10.
• Procedural color worth knowing: the Board denied Neurelis's motion for leave to seek a certificate of correction (a JRA/priority fix) via 37 C.F.R. § 42.71, finding the effort untimely; the '546 patent also saw a statutory disclaimer filed 2026-01-02 (Ex. 2075).
• Defensive value for a '400 defendant: high as intelligence, zero as estoppel. A FWD on the '546/'414/'786 does not estop anyone from attacking the '400, because § 315(e)(2) estoppel is patent- and claim-specific. But a unpatentability FWD on 2026-09-17 would hand every generic a Board-blessed roadmap on priority and motivation-to-combine for this exact family — and the '400 shares the '558 provisional lineage. Watch that date closely; it is seven days after the date of this report.

Minor discrepancy to flag against the previously generated sections: the litigation summary stated the family IPRs had "trials scheduled" for 2026-06-16. The Director's operative decision gives a projected FWD due date of 2026-09-17. These are reconcilable (2026-06-16 likely an oral-hearing/trial-setting date; 2026-09-17 the statutory FWD deadline) but the two sources should not be quoted interchangeably. Other previously generated content — filing dates, priority date, Orange Book expiration, claim 1 scope from the complaints — reconciles cleanly with what I found.

No Unified Patents (or any other defensive aggregator) filing was found in the chain on this family; the challengers are all ANDA defendants acting in their own interest.

Strategic summary

Claim status on 12,521,400. Claims 1–21 are all UNTESTED at the PTAB. None is canceled; none is sustained; none has been construed by any PTAB panel. The only PTAB claim constructions that exist in this family are for the '876, '546, '414 and '786 claims. Do not assume the '400's claim 1 (diazepam 1–20 mg, tocopherols 45–85% w/w, alcohols 25–40% w/w, 50–150 µL solution, spray formulation) will rise or fall with them — note that claim 1 does not recite dodecyl maltoside, which was a load-bearing limitation in several sibling claims and a focal point of the priority disputes. That omission cuts both ways: it may remove the priority-support vulnerability the Board has been exploiting, or it may simply shift the fight to the claimed weight ranges and the 50–150 µL volume limitation.

Estoppel landscape. As of 2026-09-10 there is no § 315(e)(2) estoppel of any kind attaching to the '400 patent — not for Padagis, not for Lupin, not for Amneal, not for Strides. Every ground a defendant would reasonably want to raise (§ 102/§ 103 on Gwozdz, Meezan, and any art a skilled searcher would find; § 112 written description and enablement on the claimed ranges and volumes) remains fully available. That window is genuinely open, but it is bounded on two sides: (i) the § 315(b) one-year bar, which runs from service of the complaint, not filing — Padagis was served 2026-04-09 (answer due 2026-04-30), so its bar date is on or about 2027-04-09; Lupin and Amneal were sued 2026-03-11 and Strides on 2026-07-31, with service dates not verified; and (ii) § 311(c), which keeps IPRs out of reach until 2026-10-13. A defendant therefore has roughly a six-month usable filing window (2026-10-13 → its own § 315(b) date) rather than a year.

The PGR point — the most under-appreciated item in this report. The PGR window closes nine months after grant, i.e., on or about 2026-10-13 — about 33 days from today. A PGR is the only AIA vehicle that can reach § 112 written description and enablement, and the Neurelis family's demonstrated vulnerability is precisely a written-description/priority problem: the Board found the '876 claims unsupported by the '558 provisional, while examiners later reached the contrary conclusion on sibling claims — the very contradiction the Acting Director called "material error." IPR cannot raise § 112. If a '400 defendant wants a § 112 attack at the PTAB, the PGR clock is the hard deadline, and it expires essentially simultaneously with the date IPRs become available. Note also that a PGR petition carries § 325(e) estoppel and that § 315(b)'s one-year bar applies to PGRs as well — so for early-served defendants the § 315(b) date is not the binding constraint; the PGR window is.

Pattern signals. The same petitioner (Padagis) has filed three parallel IPRs against this family and has now served four Paragraph IV notice letters; it is a persistent, well-capitalized challenger, and the '400 was appended to its Padagis IV action filed 2026-04-08. Neurelis litigates hard and defends its portfolio aggressively (it has also sued FDA over the related orphan-drug exclusivity question, and it filed for discretionary denial in the family IPRs). Expect a '400 IPR or PGR petition from one or more of the four defendants reasonably promptly after the gates open, and expect Neurelis to contest institution.

Institutional headwinds to price in. The environment for petitioners is materially worse than a year ago: PTAB petition filings fell 64.2% year-over-year to 131 in Q1 2026, the overall institution rate is around 30%, and bio/pharma institution rates have been reported as low as ~28% by some analysts. A 2026-03-11 Director memo added a US-manufacturing factor to discretionary denial, and there is a proposed "one and done" rule that would bar a second IPR against a patent that has already survived a validity challenge anywhere. Petitions against the '400 will be filed into a discretionary-denial-averse Board, and — per the Advanced Bionics/"material error" reasoning that carried the day in IPR2025-00464/465/466 — the strongest rhetorical hook remains the IPR2019-00451 contradiction, not raw prior-art mapping.

Recommended next steps

1. If you are a defendant served with a complaint asserting claims 1–21 of the '400: you have no estoppel to worry about, and no PTAB "record" to distinguish. Say so plainly — any assertion by the patent owner that the '400 has "withstood PTAB scrutiny" is false as of 2026-09-10.
2. Calendar 2026-09-17. The family FWDs in IPR2025-00464/465/466 are projected for that date. An unpatentability holding — particularly another priority/'558 provisional finding — is the highest-value, lowest-cost piece of evidence you can put in front of your own expert before you draft anything. Monitor at https://ptacts.uspto.gov/ and the PTAB decisions page at https://www.uspto.gov/patents/ptab.
3. Calendar 2026-10-13 as a dual deadline: (a) last day to file a PGR (the only route to § 112), and (b) first day an IPR may be filed. If § 112 / written description is part of your theory, the PGR decision must be made inside the next ~33 days and cannot be rescued later by an IPR.
4. Compute each defendant's § 315(b) date from the actual service date, not the filing date (Padagis: served 2026-04-09 → bar on or about 2027-04-09). Docket source: https://www.courtlistener.com/docket/73167119/neurelis-inc-v-padagis-llc/
5. Preserve the § 112 case for PGR and the § 102/§ 103 case for IPR, and coordinate the two. Filing an IPR first is pointless — § 311(c) blocks it until the PGR window has closed anyway.
6. Verify before filing. I could not query PTAB E2E or the ODP API directly in this session. Confirm the zero-count and the absence of any PGR by searching PTAB E2E by patent number (12,521,400) and by application number (18/782,646) before you state the negative in a filing or to a client. If a PGR has been filed in the last few weeks, it may not yet be indexed by the aggregators I searched.

Confidence: High that no IPR exists against 12,521,400 — the § 311(c) 9-month bar from the 2026-01-13 grant makes one legally impossible before 2026-10-13, independent of search coverage. Moderate-to-high that no PGR exists yet (nothing surfaced, but the window is open and recently-filed PGRs may lag in indexing). High on the adjacent family proceedings, the Director's informative decision, and the IPR2019-00451 posture. Not verified in this session: APJ panel rosters, the CAFC docket number for the '876 appeal, exact § 315(b) service dates for Lupin/Amneal/Strides, and the precise PGR-window closing date (calculated as 9 months from grant; confirm on the face of the grant notice).

I'll research the assignment record for US 12,521,400, starting with the USPTO Assignment Center and cross-referencing corporate/SEC sources.

Let me dig into the specific assignment records (reel/frame) and correspondent of record.

Important data-availability note (read first)

I was able to confirm that assignments exist for this patent, but I was not able to retrieve the reel/frame numbers, execution dates, or the correspondent of record from the USPTO Patent Assignment Center within my search budget. The Assignment Center and its legacy mirror (assignment.uspto.gov) are JavaScript-driven search interfaces that my search tooling cannot query directly, and no indexed third-party copy of the '400 assignment abstracts surfaced.

What I do have is the Google Patents legal-events ("reassignment") panel for US 12,521,400, which is populated from the USPTO assignment dataset and shows the assignor/assignee/conveyance for each recorded transfer. I am therefore reporting the chain with high confidence on parties and conveyance type, and explicitly flagging reel/frame, execution dates, and correspondent as unverified. Do not treat the absence of reel numbers below as "no records" — records exist; I simply could not pull the identifiers. Verify at:

• https://assignmentcenter.uspto.gov/ (search by patent number 12521400 or application 18/782,646)
• https://assignment.uspto.gov/patent/index.html

Inventors

Named inventors on the face of US 12,521,400, with the entity each was associated with at the time the underlying invention was made (the 2008 priority work — note the '400 is a late continuation, so "time of filing" for this specific application, 2024-07-24, is not when any of them were employed by anyone relevant):

Pattern observations:

• This is a two-employer, two-side invention. Five named inventors split across Questcor (the drug developer) and Particle Sciences (the formulation CDMO) reflect the 2007–2008 Master Service Agreement between those two companies referenced in ¶7 of the Hale Declaration. That structure — a pharma sponsor plus a formulation contractor — is what produced the dual-layer ownership cleanup described below.
• The corporate homes of the inventors did not survive intact, and that is the driver of the assignment record. Questcor was acquired by Mallinckrodt plc (2014), so no inventor remained with the "original" filer. Particle Sciences was later acquired by Lubrizol (I believe circa 2019, but I could not re-verify this within my step budget — treat as unverified). Questcor's rights were already contracted away to Hale BioPharma Ventures by the June 27, 2008 Asset Purchase Agreement (Neurelis Ex. 2031, Bates Neurelis-00218626, signed by Hale as CEO of HBV).
• Caveat on the "all inventors gone within 12 months" heuristic: the classic fire-sale precursor does not map cleanly here. The asset was sold by Questcor before any clinical or animal testing (Hale Decl. ¶5), i.e., the transfer was an early-stage carve-out, not a distressed departure. Do not read this chain as an abandonment signal.
• The presence of Hale and Maggio as assignors on the '400 record even though neither is a named inventor on the '400 is worth flagging as an anomaly. The most likely explanation is a confirmatory/omnibus assignment covering the family rather than a bare inventor-to-assignee conveyance for one application, but I cannot confirm the instrument's text. Google's reassignment panels also occasionally attribute a family-level assignment to an individual member; I cannot rule that out.

Original assignee

Entity named on the issued patent: Neurelis, Inc. (Google Patents lists Neurelis both as "Original Assignee" and "Current Assignee"; the printed assignee is Neurelis, Inc.)

• Does it ship a product embodying the claims? Yes — this is dispositive for the analysis below. Neurelis markets VALTOCO® (diazepam nasal spray; 5 mg / 7.5 mg / 10 mg per spray; NDA 211635), FDA-approved January 10, 2020, for acute treatment of intermittent, stereotypic episodes of frequent seizure activity (seizure clusters). US 12,521,400 is listed in the Orange Book against VALTOCO with an anticipated expiry of March 27, 2029. Neurelis is also the named defendant in the 2025–2026 IPRs filed by Padagis and the named plaintiff in the Delaware ANDA actions.
• Primary line of business: commercial-stage, privately held specialty pharmaceutical company (San Diego, CA) — development, regulatory approval, and commercialization of intranasal CNS/rescue therapeutics. This is not a licensing entity; its revenue model is drug sales.
• Current status: Operating. I could not complete verification of its current financing position or any change-of-control event within the search budget, so I will not state a capitalization figure. What I can confirm from the record: Neurelis filed the '400 application itself on 2024-07-24 and is funding multi-front Hatch-Waxman litigation plus PTAB defense — behavior inconsistent with a dormant or dissolved entity.
• Predecessor of record for the asset: Hale BioPharma Ventures, LLC ("HBV"), a San Diego venture builder/incubator formed by David F. Hale in June 2006 (Hale Decl. ¶3), whose stated role is forming and developing life-science companies and realizing returns "through their eventual sale or public offering." HBV is an investment vehicle, but it is a portfolio operating-investor, not an anonymous assertion shell — it had a named founder, a public-facing identity, and it incubated Neurelis specifically to take the QSC-002 program forward (Hale Decl. ¶9).

Assignment timeline

Two recorded assignments are reflected in the Google Patents reassignment dataset for US 12,521,400, both dated 2024-07-24 — the same date the continuation application 18/782,646 was filed.

Reel/frame: NOT RETRIEVED. Execution dates below are shown as "not retrieved"; the 2024-07-24 figure is the date Google Patents records for both entries and matches the application filing date. Correspondent: NOT RETRIEVED. I have no verified correspondent-of-record name for either instrument, so signal #3 below is scored conservatively.

• 2024-07-24 (executed: not retrieved) / recorded 2024-07-24 — Reel NNNNNN/NNNN (not retrieved)

  • Conveyance: Assignment — "ASSIGNMENT OF ASSIGNOR'S INTEREST"
  • Assignors: HALE, DAVID F.; LOXLEY, ANDREW; MITCHNICK, MARK; GWOZDZ, GARRY THOMAS; CARTT, STEVE; MAGGIO, EDWARD T.; MEDEIROS, DAVID
  • Assignee: HALE BIOPHARMA VENTURES, LLC
  • Correspondent: not retrieved. (Recurrence check: cannot be run. Flagging as a data gap rather than asserting non-recurrence.)
  • Context: Confirmatory chain-of-title consolidation — the 2008 priority-era rights, which had been contracted from Questcor to HBV by the June 27, 2008 APA, are formally vested in the venture entity that has held the asset since 2008. Not an acquisition and not a sale.

• 2024-07-24 (executed: not retrieved) / recorded 2024-07-24 — Reel NNNNNN/NNNN (not retrieved)

  • Conveyance: Assignment — "ASSIGNMENT OF ASSIGNOR'S INTEREST"
  • Assignor: HALE BIOPHARMA VENTURES, LLC
  • Assignee: NEURELIS, INC.
  • Correspondent: not retrieved. (See note above.)
  • Context: Internal reorganization / transfer out of the venture vehicle and into the operating company that commercializes VALTOCO. Direction of travel is investor-vehicle → product company, i.e., the inverse of a shell-entity transfer.

• No further post-issuance assignments were identified. The chain appears to terminate at Neurelis, Inc. The asserted patents are being litigated by the product owner, not by a transferee.

Related-family context (different applications — included only to show the assignee's recording practice, not as part of this patent's chain): assignments recorded against Neurelis's 2021/2022 application US20220218598A1 show individual inventor→Neurelis conveyances under reels 059797/0994 (assignor Misra, Sunita) and 059798/0076, 059798/0050, 059798/0239, 059798/0152, with named individuals including Rabinowicz, Adrian L.; Carrazana, Enrique J.; Maggio, Edward T.; and Madden, Stuart. This demonstrates Neurelis runs a routine, properly recorded, real-party inventor-assignment practice in the 059797–059798 reel block. I could not match the '400's records to a reel, and these are not the '400's records.

Prosecution-side representative (not the assignment correspondent — do not conflate): PatentGuru indexes the '400 as prosecuted through DLA Piper LLP. DLA Piper is a large general-practice firm that does both operating-company and NPE patent work; a single appearance carries no signal weight. I could not confirm whether DLA Piper or another agent filed the assignment recordations.

Timeline diagram

timeline
    title Ownership of US 12521400
    2006 : Hale BioPharma Ventures formed
    2008 : Priority application filed
         : Questcor asset acquired by HBV
    2017 : First family patent issues
    2020 : Valtoco approved by FDA
    2024 : Continuation application filed
         : Inventors assign to Hale BioPharma
         : Hale BioPharma assigns to Neurelis
    2026 : Patent 12521400 issues
         : Neurelis sues ANDA filers

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT.
The only entity in the chain with a shell-suggestive name is Hale BioPharma Ventures, LLC ("Ventures"). It does not satisfy the pattern, on three independent grounds: (a) the documented purpose is a venture builder/incubator founded in 2006 by a named individual, David F. Hale, with a public presence and a track record of forming and exiting companies (Oncternal Therapeutics, Neurelis, Biocept, Somaxon, SkinMedica) — it is an operating investor, not a single-purpose assertion vehicle; (b) it is not "licensing-only": it funded, staffed, and incubated the product program, and its principal sat as a party to the 2008 Asset Purchase Agreement; and (c) critically, the transfer runs in the wrong direction — HBV divested to the operating company rather than being the terminal acquirer. No registered-agent-service address, no anonymous single-member LLC, no products-absent finding. No reel/frame to cite because I could not retrieve them; the party-level facts above come from Neurelis Ex. 2047 (Hale Declaration, IPR2025-00464) and the Google Patents reassignment record dated 2024-07-24.

2. Known asserter in the chain — NOT PRESENT.
Assignees are Neurelis, Inc. and Hale BioPharma Ventures, LLC. Neither appears on any of the asserter lists enumerated (Acacia, Marathon, Intellectual Ventures, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, Spangenberg entities), and neither surfaced in Unified Patents or RPX high-frequency-plaintiff directories in my searches. The record confirms the opposite profile: Neurelis is the target of competitor IPRs (Padagis IPR2025-00464/-00465/-00466; Aquestive IPR2019-00451), which is the posture of an operating patent owner, not an assertion fund.

3. Repeat correspondent across the chain — UNCLEAR (cannot be scored).
This is the one signal I genuinely cannot evaluate, and I will not guess. The correspondent-of-record on both 2024-07-24 recordings was not retrievable within my search budget. There is no basis to assert either recurrence or non-recurrence. This signal must be re-run against the Assignment Center record. Note that if the two recordings do share a single correspondent, that would still be weak evidence here — two same-day, same-firm recordations in an internal reorganization are routinely handled by one attorney at the company's outside counsel, which is exactly the benign explanation.

4. Cascading transfers — NOT PRESENT.
Two assignments, both recorded 2024-07-24, i.e., a same-day two-link chain, then nothing. There is no sequence of successive LLC hops, no shared correspondent address across unrelated entity names, and no common-principal laundering. A same-day pair is the canonical signature of an internal reorganization, not of a staged cascade. (Contrast: a cascade would show HBV → HBV IP Holdings → an unrelated LLC across 12–24 months.)

5. Pre-litigation transfer — NOT PRESENT.
The assignment is dated 2024-07-24, which is (i) exactly the filing date of application 18/782,646 — the transfer is conjunctive with filing, done to establish a clean standing record at the moment of filing — and (ii) roughly 20 months before the first infringement action naming the '400 patent (D. Del. 1:26-cv-00258, filed 2026-03-11). That is well outside the 6-month window. No transfer was made to enable or time an assertion.

6. Bankruptcy fire-sale — NOT PRESENT.
No Chapter 7 or Chapter 11 proceeding by any assignor or assignee appears in the record. The 2008 transfer out of Questcor was a negotiated, pre-clinical carve-out under an Asset Purchase Agreement executed June 27, 2008 (Neurelis Ex. 2031) — the seller retained its core business (Acthar) and was later acquired by Mallinckrodt on a going-concern basis. No distressed sale of this family.

7. Privateering — NOT PRESENT.
The chain terminates at Neurelis, Inc., the company that makes and sells the patented product. No NPE has been interposed to assert on Neurelis's behalf. The enforcement activity documented in the earlier sections — Padagis IV (1:26-cv-00401), Lupin II (1:26-cv-00258), Amneal I (1:26-cv-00261), and Strides (1:26-cv-00947), each asserting claims 1–21 of the '400 — is Neurelis suing alleged ANDA infringers of its own commercial product, the textbook definition of direct operating-company enforcement.

8. Defensive aggregator — NOT PRESENT.
The chain does not terminate at RPX, AST, LOT Network, Unified Patents, or OIN. No defensive neutralization.

Supplementary observation (not one of the eight signals, but relevant to ownership strategy): This family exhibits a rolling-continuation, life-cycle-management pattern. Members issued 2017 ('876, owned by Hale Biopharma Ventures at the time of the 2019 IPR), 2022 ('414), 2023 ('786), 2025 ('664, '852, '061, '611), and 2026 (the '400) — from a single March 28, 2008 priority date, roughly tracking each competitor's Paragraph IV notice. Newly issuing continuations generate fresh Orange Book listings and fresh 30-month stays. This is aggressive brand-pharma exclusivity practice, which is legitimate and common, but it is not evidence of NPE/troll conduct — the distinction matters, because the accumulation of many patents from one priority date can superficially resemble a "portfolio aggregation" signal.

Verdict

Operating-company assertion.

The asserted chain is Hale BioPharma Ventures, LLC → Neurelis, Inc., recorded 2024-07-24 — the same day Neurelis filed application 18/782,646 — with no subsequent transfer, no transfer within 6 months of any asserted suit, no cascading LLC hops, and no NPE intermediary. Neurelis is the party shipping VALTOCO®, the Orange Book–listed product the claims cover (NDA 211635; anticipated expiry March 27, 2029), and it is litigating as plaintiff against actual generic competitors — Padagis (1:26-cv-00401), Lupin (1:26-cv-00258), Amneal (1:26-cv-00261), and Strides (1:26-cv-00947) — asserting claims 1–21 of the '400, while simultaneously defending competitor IPRs on the family. Nothing in the ownership record shows an assertion vehicle acquiring the patent for enforcement revenue; the only "aggro" feature is a rolling continuation strategy, which is a brand-pharma practice, not an NPE indicator.

Two explicit caveats on this verdict:

1. Reel/frame numbers, execution dates, and the correspondent of record were not retrieved. Signal #3 (repeat correspondent) is therefore unscored, and the timeline above rests on the Google Patents reassignment dataset (USPTO-derived) plus Neurelis Ex. 2047 rather than on my own read of the Assignment Center instruments. Verify at https://assignmentcenter.uspto.gov/ and https://assignment.uspto.gov/patent/index.html by searching patent number 12521400 or application 18/782,646.
2. The '400's assignment is a family-level cleanup, not the origin of title. Title traces to the June 27, 2008 Asset Purchase Agreement between Questcor and HBV. If you need the complete chain of title for standing purposes (e.g., for a § 271 action or an IPR), you must pull the 2008-era HBV instruments as well — the two 2024-07-24 recordings are confirmatory and will not by themselves establish an unbroken record from Questcor and Particle Sciences.

I'll search for the USPTO record of US 12,521,400 and its cited prior-art references.

Now let me find the specific references cited on the face of the '400 patent.

1. Verification step — what I could and could not confirm

Confirmed: US 12,521,400 B2, "Administration of benzodiazepine compositions," application US 18/782,646, filed 2024-07-24, issued 2026-01-13, assignee Neurelis, Inc., inventors Cartt; Medeiros; Gwozdz; Loxley; Mitchnick. Source: https://patents.google.com/patent/[US12521400B2](/patent/US12521400B2)/en. The full claim text (claims 1–21) is reproduced at https://www.drugpatentwatch.com/p/patent-claims/[12521400](/patent/12521400).

Not retrievable within my search budget — this is the central limitation on this deliverable: I could not open the patent's own face-of-patent "References Cited" / "Patent Citations" table (Google Patents "Patent Citations" section, USPTO Patent Center full text, or the front page of the patent PDF). Google's indexed snippets for US12521400B2 returned the description and definitions, not the citations table, and my later queries hit the step limit. Therefore I cannot state, as verified fact, which references appear on the face of the '400 patent. Everything below is provenance-labeled:

• [SPEC] = cited inside the '400 specification itself (these are almost certainly the references listed in the patent's "References Cited" section).
• [FAMILY-IPR] = the prior art actually relied on in the co-family PTAB proceedings and appeals (same specification, same priority chain), i.e., the art the challengers treat as the operative art for this family.
• [CITATION-NETWORK] = references surfaced in the citation/cited-by tables of related patents — not verified as being on the '400's face.

Contradictions with the previously generated sections (flagged as requested):

1. The prior summary described claim 1 as requiring diazepam "fully dissolved." The claim text per DrugPatentWatch reads "diazepam dissolved in about 45% to about 85% (w/w)…" — "fully" appears to be the complaint's characterization, not claim language. Treat "fully dissolved" as unverified.
2. The prior summary inferred that additional independent method-of-treatment claims likely exist. On the claims text I retrieved, claim 12 is a second independent composition claim ("A pharmaceutical composition for intranasal administration comprising:"), and claims 1–18 read as composition claims. No method claim was visible in what I could retrieve; claims 19–21 were truncated in my source. Earlier inference should be downgraded.
3. Date discrepancy: the task states the current date is April 26, 2026, but my system date is 2026-09-10 and several retrieved sources are dated May–August 2026 (e.g., DrugPatentWatch "Last Updated: August 7, 2026"). I have treated the later-indexed material as ground truth per the operating rules.

Claim 1 as issued (DrugPatentWatch): about 1–20 mg diazepam dissolved in about 45–85% (w/w) tocopherol(s)/tocotrienol(s) (α/β/γ/δ-tocopherol, α/β/γ/δ-tocotrienol, tocophersolan, any esters, combinations) and about 25–40% (w/w) of one or more alcohols (ethanol, propyl, butyl, pentanol, benzyl), wherein the composition is a solution of 50 µL to 150 µL, and is in a pharmaceutically-acceptable spray formulation for intranasal administration. Claims 5 (10–250 mg/mL), 6 (60–75% tocopherol), 7 (stabilizers/pH adjusters/preservatives), 8–9 (absorption enhancer, up to 20% w/v), 10 (~100 µL), 11 (2–10 mg) depend from claim 1; claims 13–18 depend from claim 12.

2. The legal gate that dominates the §102 analysis

Because claim 1 is a unit-dose claim, the §102 question is mostly a date question, not a disclosure question:

• Pre-AIA vs. AIA: The '400 claims a chain to 2008-03-28, so pre-AIA §§102/103 presumptively apply. But the '400 is a 2024-filed continuation; if any issued claim lacks written-description support in the pre-2013-03-16 chain, the AIA (§102(a)(2), §102(b)(2)(C)/§103) framework governs instead. This is contested in the family: in IPR2019-00451 the Board held the parent '876 patent's claims were not entitled to the '558 provisional's 2008-03-28 date (insufficient written description for the "alkyl glycoside" limitation), which made the 2009-filed family disclosures §102(e)(1) art (Federal Circuit briefing quoting the Board: "The '558 Provisional Fails To Provide Adequate Written Description Support… And Gwozdz And Cartt'784 Are Prior Art Under 35 U.S.C. § 102(e)(1)").
• Crucially, claim 1 of the '400 does not recite an alkyl glycoside. The priority defect that killed the '876/'414 claims may therefore not carry over. If the '400's claims keep the 2008-03-28 date, then the 2009 Gwozdz disclosure is not prior art at all and Meezan alone (§102(b), 2006) is insufficient. This is the single most important §102 issue for the '400.
• Also live: whether the Gwozdz/Cartt art is "by another" (inventive entity differs) and whether pre-AIA §103(c) common-ownership disqualification applies. The Board has already treated the art as available, but for the '400 it should be re-tested.

3. Prior-art references, with §102 mapping

A. The operative art for this family

A1. "Gwozdz" — WO 2009/120933 (closest art) [FAMILY-IPR]

• Citation: WO 2009/120933 (PCT), naming Garry Thomas Gwozdz — a named inventor of the '400. Filed 2009-03-27 (per the POPR's reference to "the March 27, 2009 Gwozdz PCT application"); published 2009. (I did not independently verify the exact publication date; the IPR record refers to it as "Gwozdz (WO 2009/120933)" and cites it in column format — Ex. 1009 — suggesting the US counterpart was also used.)
• Disclosure (per the IPR record): pharmaceutical solutions for nasal administration containing a benzodiazepine — expressly diazepam at ~8–10% (w/w); tocopherols/tocotrienols ("tocols or vitamin E") in 30–99% (w/w), more narrowly 60–99% and 70–90% (w/w); alcohols in 1–70% (w/w) with ethanol 10–30% (w/w) and benzyl alcohol specifically called out; a tocopherol:ethanol ratio list of 95:5 … 60:40; optional "penetration enhancers"; alpha-tocopherol as a preferred tocopherol.
• §102 assessment: This is the only identified reference that discloses the claim-1 chemical combination plus the intranasal-solution form. It nonetheless does not appear to anticipate claims 1–21, for three independent reasons:

1. The 50–150 µL unit-volume limitation — no disclosure identified in the record of a single 50–150 µL dose volume. Every claim in the set I retrieved depends from claim 1, so all would stand with it. (Residual risk: if independent claim 12 omits the volume limitation — its text was truncated in my source — claim 12 becomes the leading §102 candidate against Gwozdz and must be checked. This is my #1 verification task.)
2. The absolute dose limitation (1–20 mg) — Gwozdz teaches a concentration (% w/w), not a 1–20 mg unit dose.
3. Range specificity — Gwozdz's 30–99% tocopherol and 1–70% alcohol ranges encompass but do not describe the claimed 45–85% and 25–40% sub-ranges; an encompassing broader range is a §103 (In re Peterson) issue, not anticipation.

• Realistic role: §103 primary reference. But its availability as prior art depends entirely on the priority-date question in §2 above.

A2. Meezan — U.S. Pub. No. 2006/0046962 A1 [FAMILY-IPR]

• Citation/dates: US 2006/0046962 A1 (Application No. 2006/0046962 as cited in the IPR record), published 2006-03-02; filing in 2005 (exact filing date not verified in this session).
• Disclosure: teaches alkyl glycosides, including n-dodecyl β-D-maltoside (dodecyl maltoside), as intranasal absorption enhancers to increase bioavailability.
• §102 assessment: Published more than one year before 2008-03-28, so it is §102(b) art regardless of the priority outcome. It cannot anticipate any of claims 1–21 — it discloses neither diazepam in a tocopherol/alcohol vehicle, nor the 45–85%/25–40% matrix, nor a 50–150 µL unit dose. It is relevant only to claims 8–9 (absorption enhancer, up to 20% w/v) and as the §103 secondary reference, exactly as in the family IPRs.

A3. "Cartt '784" [FAMILY-IPR]

• Citation: referred to in the family's PTAB/Federal Circuit briefing as "Cartt'784," treated as §102(e)(1) art; the briefing cites it at ¶¶47–48, 124–126, 128–129, 250–252 and notes it expressly identifies alkyl glycosides including n-dodecyl β-D-maltoside and describes dosing regimens for particulate formulations.
• What I could not do: resolve "Cartt '784" to a complete citation (patent number or publication number) within my search budget. I will not guess at the number. This is a gap; it should be resolved from the IPR exhibit list / appellant brief in the '876 appeal.
• §102 assessment: Its disclosure is, by the briefing's own account, broader than and overlapping with the '400's disclosure (it is described as containing language "almost verbatim" to the '558 provisional). If it discloses a specific 50–150 µL intranasal diazepam solution of 1–20 mg in tocopherol/alcohol, it could potentially anticipate claim 1; if it discloses only ranges and particulate dosing regimens, it is §103 art. Unresolved pending the actual text.

B. Additional candidate art surfaced in the citation network (verify before relying)

C. References cited in the '400 specification itself [SPEC]

These are the citations most likely to appear on the patent's face, and they are answered claim-by-claim below:

• US 5,661,130 — alkyl glycoside penetration enhancers (Intravail®-type chemistry), incorporated by reference for the definition of "alkyl glycoside." 1990s vintage; exact issue date not verified. Cannot anticipate claims 1–21 (it is an enhancer-chemistry reference; no benzodiazepine/vitamin-E solution, no unit dose). Relevant only to claims 8–9.
• US 6,193,985 — Vitamin E TPGS (tocopherol PEG succinate), incorporated by reference. 2001; exact date not verified. Notably, the '400 specification expressly excludes Vitamin E TPGS from the solution embodiments — i.e., the patentee cites it as the teaching to be avoided. Cannot anticipate claim 1, which requires free tocopherol/tocotrienol (incl. esters such as tocophersolan), not the PEG-succinate conjugate.
• Benzodiazepine manufacturing-process patents cited in the specification: US 3,987,052 (alprazolam); US 3,371,085; US 3,109,843; US 3,136,815; US 3,102,116 (diazepam); US 3,567,710 and US 3,299,053 (flurazepam); US 3,296,249 (lorazepam); US 3,243,427 (medazepam); US 3,722,371 (mexazolam); US 4,280,957 and US 5,831,089 (midazolam); US 3,340,253 and US 3,374,225 (temazepam). (These are 1960s–1990s compound/process patents — Hoffmann-La Roche, Upjohn and related originators. I did not verify individual issue dates in this session; the '400 specification cites them only as synthetic routes.) None can anticipate claims 1–21: they are methods of making the active ingredients, and a compound-synthesis reference does not describe a nasal pharmaceutical composition, the tocopherol/alcohol matrix, or a 50–150 µL unit dose. Their only significance is that they are the specification's own §102(b)-vintage citations establishing the benzodiazepines as known.
• SIGMA Catalog, 1988, pp. 316–321 (biological detergents) — cited in the specification. §102(b) printed publication, but it is a detergent catalog and was held insufficient as written-description support for "alkyl glycoside" in the family IPR. Cannot anticipate any claim.

4. Bottom line on §102

1. No reference I could identify anticipates claims 1–21 on the present record. Claim 1's combination of (i) 1–20 mg diazepam, (ii) 45–85% w/w tocopherol/tocotrienol, (iii) 25–40% w/w alcohol from the recited list, (iv) a 50–150 µL solution, and (v) a nasal spray formulation is not disclosed as a unit dose by any reference I can identify. Gwozdz is the closest and still fails on the volume and absolute-dose limitations and on range specificity — so it is a §103 primary reference, not a §102 reference.
2. The only realistic §102 path runs through the priority date, not through a better reference. If the '400's claims are not entitled to 2008-03-28 (as the Board held for the parent '876 claims), then the March 27, 2009 Gwozdz PCT/Cartt '784 disclosures become §102(e)(1) art, and any of them that discloses a 50–150 µL intranasal diazepam solution would anticipate claim 1 and all claims depending from it. Because claim 1 omits the alkyl-glycoside limitation that defeated the parent claims, that path is narrower for the '400 than it was for the '414/'786 patents — but it is not closed.
3. Highest-value verification targets, in order: (a) the '400's actual face-of-patent citation list (USPTO Patent Center / Google Patents "Patent Citations"); (b) the full text of claim 12 (does it recite the 50–150 µL limitation?); (c) full citation and disclosure of "Cartt '784" (IPR exhibit list / '876 appeal brief); (d) WO 2009/046444 A2 against claim 1; (e) whether any claim of the '400 lacks pre-2013 written-description support (which chooses the AIA vs. pre-AIA §102 statute).

Confidence: High on the claim-1 text, on the Gwozdz and Meezan disclosures and their use in the family IPRs, and on the §102 date analysis. Moderate on the citation-network references' relevance (their presence on the '400's face is unverified). Low/unverified and explicitly flagged: the '400's own "References Cited" list, the full text of claims 12 and 19–21, the complete citation for "Cartt '784," and the individual issue dates of the specification-cited synthesis patents. I could not complete these verifications within the available search steps; the authoritative sources are the USPTO Patent Center record for 12,521,400 and the PTAB exhibits in IPR2025-00464/-00465/-00466 (see https://ai-lab.exparte.com/case/ptab/IPR2025-00465/doc/9 and the POPR/POR documents at https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1557269](/patent/1557269) and /1557270).

I'll first try to retrieve the "Prior Art" section content for this patent, since the copy of the patent text supplied to me ends before any prior-art reference listing (it truncates mid-sentence in the "enhancer" passage).

I now have substantial data on the family's prior art. Let me verify the identities and disclosures of the key references, and get the full claim set.

Obviousness Analysis — U.S. Patent No. 12,521,400 (35 U.S.C. § 103)

Analyst note on date and source integrity — please read first

1. The "Prior Art section" was not actually present in the page text I was given. The patent copy supplied to me (fetched from the Google Patents record) ends mid-sentence in the "enhancer" discussion and truncates before any "References Cited," "Similar Documents," or "Cited By" block. The only prior-art content in that document is the sidebar field Prior art keywords: seizure; pharmaceutical composition; benzodiazepine; drug; combinations and Prior art date 2008-03-28. I therefore could not "use the results from the Prior Art section of this page" as instructed, because that section was not delivered. I state this rather than inventing a reference list.
2. What I used instead: live search results (retrieved today) covering (a) the '400 patent's own claim text as published by DrugPatentWatch, (b) the prosecution/PTAB record of the same family — in particular Padagis's IPR2025-00465 against the sibling '414 patent and Aquestive's IPR2019-00451 against the parent '876 patent — which supplies the actual prior-art references this family has been attacked with, and (c) the '400 specification's own admissions. All citations are to those sources. This is family-level prior-art record, not the '400 patent's face citations.
3. Date inconsistency (flagged per instructions): the header of this task says Current Date: April 26, 2026, the session metadata says 2026-09-10, and my search results carry timestamps of August 7, 2026 and describe a July 31, 2026 complaint (Strides) that also appears in the previously generated section. The previously generated litigation summary is thus inconsistent with an April 26, 2026 "current date." I do not resolve this; I simply flag it.
4. Contradiction with the previously generated summary: that summary characterized claim 1 as reciting "about 1 to about 20 mg of diazepam (or a pharmaceutically acceptable salt)… fully dissolved." The claim text as published does not contain "pharmaceutically acceptable salt" or the word "fully." Verbatim (DrugPatentWatch, https://www.drugpatentwatch.com/p/patent-claims/12521400):

"1. A pharmaceutical composition for intranasal administration comprising: about 1 to about 20 mg of diazepam dissolved in about 45% to about 85% (w/w) of one or more natural or synthetic tocopherols or tocotrienols selected from the group consisting of α-tocopherol, β-tocopherol, γ-tocopherol, δ-tocopherol, α-tocotrienol, β-tocotrienol, γ-tocotrienol, δ-tocotrienol, tocophersolan, any esters thereof and any combinations thereof, and about 25% to about 40% (w/w) of one or more alcohols selected from the group consisting of ethanol, propyl alcohol, butyl alcohol, pentanol, benzyl alcohol, and any combinations thereof, wherein the pharmaceutical composition is a solution of 50 μL to 150 μL, and wherein the pharmaceutical composition is in a pharmaceutically-acceptable spray formulation for intranasal administration."

Dependent claims 2–11 and 12–18 are available (claims 19–21 were truncated in the source and I could not verify them).

1. What claim 1 does not require — this is decisive for § 103

Unlike every previously litigated claim in this family ('876, '546, '414, '786), claim 1 of the '400 patent contains no alkyl glycoside / absorption-enhancer limitation and no "consisting of" closure. The enhancer appears only as an optional element (claim 8: "further comprises an absorption enhancer"). Consequently, the ground that has twice been held sufficient to invalidate the parent claims — Gwozdz in view of Meezan'962 — applies to '400 claim 1 a fortiori, because claim 1 omits one of the two elements that had to be supplied by a secondary reference.

2. Primary reference (as identified in this family's own proceedings)

Gwozdz — WO 2009/120933 (inventors Gwozdz, Loxley & Mitchnick; applicant Particle Sciences, Inc.; US provisional 61/040,281 filed 2008-03-28; PCT filed 2009-03-27; published 2009-10-01). Per the IPR record it discloses:

Sources: https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1522097](/patent/1522097)/... (Petitioner's reply, "Prior Art References Relied On By Petitioner," quoting Gwozdz, Exhibit 1014); https://ai-lab.exparte.com/case/ptab/IPR2025-00465/doc/1026.

Two elements of claim 1 are not squarely in Gwozdz: (a) the 1–20 mg diazepam dose, and (b) the 50–150 µL solution volume. Both are classic "result-effective variable / routine optimization" limitations, and the '400 specification itself supplies the admission that defeats them.

3. The volume and dose limitations are supplied by routine optimization and the patent's own admissions

The '400 specification states:

• "Mucosal membrane preparations are generally administered in metered sprays having volumes of less than 250 µL, preferably less than 150 µL, and ideally from 25 to 100 µL" (admission that sub-150 µL sprays were the conventional practice).
• "alprazolam … may be administered in 25 to 250 µL metered sprays … 50 to 150 µL, especially about 100 µL, metered sprays" — identical language for diazepam, flurazepam, lorazepam, medazepam, mexazolam, midazolam and temazepam.
• "the dosage volume ranges from about 10 µL to about 200 µL."
• Diazepam dosing: "about 1 to about 20, preferably about 2 to about 10 mg per dose."

Once Gwozdz supplies the solubilised diazepam carrier and the specification admits both the conventional metered-spray volume window and the therapeutic dose window, selecting 50–150 µL is arithmetic dictated by the dose and the drug concentration (e.g., 5 mg per 100 µL = 50 mg/mL; 10 mg per 100 µL = 100 mg/mL — squarely inside claim 5's 10–250 mg/mL). In re Aller / In re Peterson / In re Woodruff / Galderma v. Tolmar (difference in degree, not kind).

4. Combination A (the core § 103 ground): Gwozdz + Meezan'962 + the metered-dose-spray/dosing art

Meezan — US 2006/0046962 (published 2006-03-02, i.e., before the earliest asserted priority date) teaches using alkyl glycosides, specifically n-dodecyl β-D-maltoside, as intranasal absorption enhancers, and teaches that the carrier "can be … aqueous or non-aqueous … for example alcoholic or oleaginous … selected from substances such as … benzyl alcohol" (quoted at ¶113 of the rebuttal declaration at https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1557271](/patent/1557271)/...).

Cartt'784 — Application No. 2008/0279784 (per the IPR2019-00451 record) discloses dosing regimens for intranasal benzodiazepine formulations; the '876 IPR record includes expert testimony that "the dosing regimens of Cartt'784 for particulate formulations would be applicable to a solution formulation of benzodiazepines." This is the reference a petitioner would use to close the "1–20 mg / 50–150 µL per spray" gap for a composition claim reciting a dispensed volume.

Motivation to combine (KSR rationales 1–7 all apply):

1. Same field, same problem, same solution mechanism. Gwozdz (nasal vitamin-E/alcohol benzodiazepine solution) and Meezan (intranasal absorption enhancement) address the identical problem the '400 patent identifies as unsolved in 2008: diazepam's poor water solubility and slow, first-pass-limited oral rescue therapy. There is no field-crossing; the references are combinable "prior art elements according to known methods to yield predictable results."
2. Gwozdz itself directs the combination by suggesting "penetration enhancers" and other excipients — a teaching-away-free, express motivation.
3. Predictable results / reasonable expectation of success. A POSITA would expect (i) enhanced absorption from a known enhancer, and (ii) that the 1–20 mg dose and ≤150 µL spray volume follow from the therapeutic dose and nasal-cavity capacity. The IPR record contains extensive evidence that intranasal diazepam/midazolam PK had already been characterised, that benzyl alcohol was known as an antimicrobial/preservative co-solvent, and that Florida EMS was already administering injectable diazepam intranasally for seizures (Wermeling IV, Ex. 1061, cited in the IPR record).
4. Design incentives / market forces ("KSR rationale 6"): a caregiver-administered, needle-free, fast-onset rescue product for seizure clusters.

No criticality or unexpected results: the PTAB in the '876 FWD "found there was no criticality or unexpected results as to the amounts of tocopherols … and ethanol and benzyl alcohol," and the EPO file history for the counterpart application states that "No unexpected effect whatsoever has been demonstrated … for the selection of the claimed individual amounts of ethanol and benzyl alcohol" and that a skilled person "would consider adjusting the amounts … by mere routine experimentation and, by doing so, would inevitably arrive at the claimed amounts, without exercising inventive skill" (EP FH Ex. 1040, p. 0298). The '400's 25–40% alcohol range sits inside Gwozdz's 1–70% range, so the prima facie case is even stronger than in the parent.

5. Obviousness of the dependent claims (2–21, to the extent text is available)

• Claims 2–4 — species of tocopherol/alcohol already named in Gwozdz (α-tocopherol; ethanol, benzyl alcohol).
• Claim 5 (10–250 mg/mL) and claim 6 (60–75% tocopherol) — within the ranges the '400 specification itself presents as routine, and within Gwozdz's ranges; the same "no criticality" findings apply.
• Claim 7 (stabilizers, pH adjusters, preservatives, etc.) and claims 8–9 (an absorption enhancer "up to 20% (w/v)") — conventional excipients; Meezan teaches enhancers; the "up to 20% w/v" figure matches the enhancer concentrations discussed in the family's own specification (U.S. Pat. No. 5,661,130 alkyl glycosides; U.S. Pat. No. 6,193,985 Vitamin E TPGS).
• Claim 10 ("about 100 μL") — the specification itself recites ~100 µL metered sprays as the working example; the prior-art nasal spray devices are dosed in that window.
• Claim 11 / 14 (2–10 mg) — the specification's own "preferably about 2 to about 10 mg" for diazepam.
• Claim 12 et seq. — the same mapping applies; the truncated text suggests a parallel composition claim (possibly with benzyl alcohol/ethanol specifics).
• Claims 19–21 — unverified. If, as the specification's structure suggests, they are method-of-treatment claims ("administering to one or more nasal mucosal membranes…"), they rise or fall with the composition and would additionally be met by the known intranasal anticonvulsant administration practice (Florida EMS; intranasal midazolam/benzodiazepine literature).

6. The threshold issues that decide whether this § 103 case survives

This is where the analysis honestly cannot be concluded from the material I have:

1. Priority. The '400 claims 2008-03-28. Gwozdz's pre-AIA § 102(e) date is its international filing date, 2009-03-27 (English-language PCT designating the US), not 2008-03-28. If '400 claim 1 is entitled to 2008-03-28 (pre-AIA § 102/§ 103 apply), Gwozdz is not prior art at all and Ground A collapses — a challenger would then have to build the case from art published before 2008-03-28 alone (Meezan'962, 2006-03-02, is available; US 5,661,130; US 6,193,985; the 1988 SIGMA catalogue discussion in the '558 provisional; and any pre-2008 intranasal-diazepam literature).
2. The parent cases' priority attack does not transfer cleanly. Padagis's successful theory was that the '558 provisional lacked written description support for "alkyl glycoside"/"n-dodecyl β-D-maltoside," which was first added in the 2009-03-27 '439 application. '400 claim 1 recites no alkyl glycoside, so that specific theory does not disqualify claim 1. A challenger would need a different priority attack aimed at the elements claim 1 does recite (the 1–20 mg dose, the 50–150 µL solution, the 25–40% alcohol subrange). I have not seen the '558 provisional and cannot assess that.
3. Inventor overlap / § 103(c). This is unusual and important: the '400's named inventors are Steve Cartt, David Medeiros, Garry Thomas Gwozdz, Andrew Loxley and Mark Mitchnick, and the Gwozdz reference names Gwozdz, Loxley and Mitchnick. Three of the five '400 inventors are the authors of the primary reference. Under In re Land, differing (even partially overlapping) inventive entities are "by another," so § 102(e) is not automatically defeated; but pre-AIA § 103(c) disqualifies § 102(e)/(f)/(g) art for § 103 purposes if, at the time the invention was made, the art and the claimed invention were commonly owned or subject to an obligation of assignment to the same person. Whether that condition existed (Particle Sciences, Inc. vs. Hale BioPharma/Neurelis) is a fact question I cannot resolve. A challenger must address § 103(c) squarely, and this is a credible Neurelis defense.
4. Whether "Cartt'784" (App. No. 2008/0279784) is prior art at all to the '400 — same-family/applicant's-own-work questions (identical inventive entity, common ownership, § 103(c)) apply to it and to any other Neurelis-family publication (e.g., US 2009/0258865, the publication of the '439 application, used as an exhibit in IPR2019-00450).
5. Objective indicia. Neurelis will argue long-felt need, failure of others, unexpected results and commercial success (VALTOCO®), and nexus. The '876 FWD, on the record as quoted in the Federal Circuit briefing, found the secondary-consideration evidence "broadly directed to an intranasal benzodiazepine formulation rather than to the specific formulation recited in the claims" and required a nexus that was not shown. For a claim as broad as '400 claim 1 (no enhancer, no closed "consisting of"), a nexus showing will be harder still — but I have not verified the outcome of that appeal.

7. Bottom line

Technically, claim 1 of the '400 patent presents a strong § 103 case on the merits. Its carrier elements (45–85% w/w tocopherol; 25–40% w/w alcohol from the recited Markush group) fall literally inside Gwozdz's disclosed ranges (30–99% and 1–70%), for the same drug (diazepam) in the same dosage form (nasal solution/spray); the alkyl-glycoside limitation that Gwozdz lacked (and that Meezan supplied) is absent from claim 1; and the only added limitations (1–20 mg, 50–150 µL solution) are routine-optimization parameters the '400 specification itself treats as conventional and for which the family's own PTAB/EPO record contains explicit "no criticality / no unexpected results" findings. The corresponding combination a challenger would plead is:

Gwozdz WO 2009/120933 alone (for carrier, drug, ranges and spray formulation), or Gwozdz + Meezan US 2006/0046962 (enhancer, for claims 8–9/17–18) + Cartt'784 / the intranasal metered-dose-spray volume art (for 1–20 mg and 50–150 µL), with the '400 specification's admissions and the PTAB/EPO no-criticality findings as supporting evidence.

But the case is gated by priority, not by chemistry. If '400 claim 1 keeps its 2008-03-28 effective filing date, Gwozdz (PCT date 2009-03-27) drops out as prior art under pre-AIA § 102(e) and the § 103 case must be rebuilt from pre-2008 art (Meezan'962 published 2006-03-02 is the obvious candidate, but standing on Meezan alone would require it to disclose the vitamin-E/alcohol solution rather than merely an enhancer — I have not verified that). If claim 1 loses the 2008 date, the challenger must also show Gwozdz's date precedes whatever fallback date the claims receive (the '439 application appears to have been filed 2009-03-27, the same day as the Gwozdz PCT — a same-day problem a petitioner must address), and must overcome any § 103(c) common-ownership argument arising from the three-inventor overlap.

Confidence levels. High: the claim 1 text; Gwozdz's disclosed ranges and ingredients; Meezan'962's teaching of alkyl glycoside intranasal enhancers; the '400 specification's own admissions about spray volumes and doses; the existence of the "no criticality"/EPO routine-optimization findings; the Gwozdz/shared-inventor facts. Moderate: the exact § 102(e) date treatment and the § 103(c) analysis. Low/unverified: the priority support in the '558 provisional; the text of claims 19–21; the identity/date/assignee of U.S. 8,911,751 ("Compositions for nasal delivery," which described an intranasal diazepam formulation containing Vitamin E, ethanol and propylene glycol together with phospholipid and water — a candidate secondary reference whose publication date I could not verify); and whether any post-issuance IPR has yet been filed against the '400 patent (none was found; Google Patents' litigation flags list only the Delaware actions). This is a technical analyst's opinion, not a legal conclusion; a full § 103 opinion would require the '558 provisional, the '400 prosecution history, the '400's face citations, and the defendants' invalidity contentions.